ask_google
Read-onlyIdempotent
Retrieve up-to-date answers about versions, releases, API changes, and breaking news by combining Gemini with Google Search grounding.
Instructions
Gemini with Google Search grounding. Use for current/latest facts that post-date your training: versions, releases, API changes, changelogs, breaking news, on-demand web research. Do not use for stable syntax or knowledge already in your training. Short lookups or multi-paragraph briefs both work.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| model | No | Google model. 'auto' (default, recommended) picks the right tier automatically. Override with 'flash' or 'flash-lite' only if you need a specific one. | auto |
| query | No | Alias for `question`. Accepted for compatibility with callers that use the name `query`; prefer `question`. Do not set both at once. | |
| question | No | Your question for the AI researcher. Short lookups or multi-paragraph research briefs both work. Prefer 'current/latest/as of today' over hardcoding dates unless a specific historical year matters. `query` is accepted as an alias. |
Output Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| answer | Yes | ||
| sources | No | ||
| supports | No | ||
| diagnostics | No | ||
| search_queries | No | ||
| grounding_status | No | How thoroughly the answer is grounded. 'grounded' = sources + per-claim supports. 'sources_only' = pages retrieved but no per-claim mapping. 'no_sources' = search ran but returned nothing — answer is from training data, treat with high skepticism. 'not_attempted' / 'unavailable' = even worse. | |
| answer_with_citations | No |